# TIGER-AI-Lab/TheoremExplainAgent

Official Repo for "TheoremExplainAgent: Towards Video-based Multimodal Explanations for LLM Theorem Understanding" [ACL 2025 oral]

Repository: https://github.com/TIGER-AI-Lab/TheoremExplainAgent
Canonical: https://ross.abutalabs.com/products/theoremexplainagent
Homepage: https://tiger-ai-lab.github.io/TheoremExplainAgent/
Language: Python
License: MIT
License Family: permissive
Topics: llm-agents, manim, manim-animations, manim-video, rag
Last push: 2025-07-27T02:56:48+00:00

## Health v2 (maintenance only)
Score: 35/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 33, release rhythm 35, longevity 40
- inputs: {"age_days": 563, "days_push": 402, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1499, forks 200 (observed 2026-08-28T04:04:54.146582+00:00)

## What it is
An agentic AI system that generates long-form (5+ minute) Manim animation videos explaining mathematical and STEM theorems using LLM agents with RAG-based planning. It accompanies the ACL 2025 paper and includes TheoremExplainBench, a 240-theorem benchmark with automated evaluation metrics.

## Use cases
- generate animated videos explaining math theorems
- evaluate how well LLMs understand theorems
- benchmark multimodal reasoning of language models
- create manim animations automatically with an LLM agent
- research multimodal explanations that reveal reasoning flaws
- get baseline theorem explanation videos for research

## When to choose
- you need automated visual explanations of STEM theorems
- you are researching multimodal LLM reasoning and evaluation
- you want to generate long-form educational Manim videos programmatically

## When to avoid
- you need a polished end-user video creation tool rather than research code
- you want short, quick animations without agentic planning overhead
- you lack access to capable LLM APIs like o3-mini

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, rag, llm-inference, video-processing, machine-learning, prompt-engineering
- domain: artificial-intelligence, large-language-models, education
- platform: python
- tags: manim, theorem-explanation, multimodal, benchmark, stem-education, research-paper, video-generation, ai-agents, natural-language-processing, linux, macos

## Member repositories
- TIGER-AI-Lab/TheoremExplainAgent (main) score 35

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:54.146582+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T04:33:01.866612+00:00, confidence not recorded.
  - readme: https://github.com/TIGER-AI-Lab/TheoremExplainAgent (fetched 2026-08-28T04:04:54.146582+00:00, sha 8c76b3c29e1e)
  - homepage: https://tiger-ai-lab.github.io/TheoremExplainAgent/ (fetched 2026-08-29T11:38:07.583771+00:00, sha 8a39e47d0817)
- Data as of 2026-08-30T08:39:29.467469+00:00.